Genetic basis for the structural difference between Streptococcus pneumoniae serotype 15B and 15C capsular polysaccharides

Abstract

In a search for the genetic basis for the structural difference between the related Streptococcus pneumoniae capsular serotypes 15B and 15C and for the reported reversible switching between these serotypes, the corresponding capsular polysaccharide synthesis (cps) loci were investigated by keeping in mind that at the structural level, the capsules differ only in O acetylation. The cps locus of a serotype 15B strain was identified, partially PCR amplified with primers based on the related serotype 14 sequence, and sequenced. Sequence analysis revealed, among other open reading frames, an intact open reading frame (designated cps15bM) whose product, at the protein level, exhibited characteristics of previously identified acetyltransferases. Genetic analysis of the corresponding region in a serotype15C strain indicated that the same gene was present but had a premature stop in translation. Closer analysis indicated that the serotype 15B gene contained a short tandem TA repeat consisting of eight TA units. In serotype 15C, this gene contained nine TA units that resulted in a frameshift and a truncated product. Genetic analysis of 17 serotype 15B and 15C clinical isolates revealed a perfect correlation between the serotype and the length of the short tandem repeat in the putative O-acetyltransferase gene. The number of TA repeating units varied between seven and nine in the various isolates. Together, the data strongly suggest that the structural difference between serotypes 15B and 15C is based on variation in the short tandem TA repeat in the O-acetyltransferase gene and that the transition between serotypes is due to slipped-strand mispairing with deletion or insertion of TA units in the cps15bM gene.

FIG. 1.

Diagram showing the structures of the repeating units of the capsular polysaccharides of S. pneumoniae serotypes 14, 15B, and 15C (14, 22). + 0.7 OAc* indicates that on avareage 70% of the repeating units is O acetylated. The site of the O acetylation is unknown (14), although Venkateswaran et al. (51) suggested that it is linked to a galactose moiety.

FIG. 2.

Comparison of the organizations of the cps loci of S. pneumoniae serotypes 14 and 15B. The genes that are present in the loci are indicated by different arrows, as follows: solid arrows, 5′ conserved genes; arrows with horizontal stripes, glycosyltransferases; arrow with horizontal dashes, polymerase; arrows with vertical dashes, repeating unit transporters; arrows with vertical stripes, nonsugar transferases; arrows with tessellated pattern, nucleotide sugar biosynthesis genes; arrows with dots, genes with unknown functions. The dashed arrows represent genes that were previously suggested to be present in serotype 15B (19). The small open arrow is not involved in CPS synthesis in serotype 14. The putative promoters of the loci are indicated by bent arrows. The putative genes cps14A to -K and cps15A to -Q are labeled A to K and A to Q, respectively.

FIG. 3.

Comparison of the nucleotide sequences at the 5′ ends of the cps14K and cps15bK genes. The amino acid sequences encoded by the ORFs are shown above the sequence for cps14K and below the sequence for cps15bK. Start codons of the ORFs are underlined. Putative ribosome binding sites are indicated by dashed underlining. The arrow indicates the position of the deletion in serotype 14.

FIG. 4.

Transmembrane segment predictions for putative O-acetyltransferases. The predicted profiles were generated for S. pneumoniae Cps15bM (serotype 15B), Cps9vO (serotype 9V), Cap33fM (serotype 33F), and an acetyltransferase of P. putida by using the program DAS (http://au.expasy.org). There are two cutoffs indicated on the plot. The strict cutoff (represented by the solid line) at a DAS score of 2.2 is informative in terms of the number of matching segments. The loose cutoff (represented by the dashed line) gives the actual location of the transmembrane segment.

FIG. 5.

Schematic representation of the putative O-acetyltransferase genes, showing the locations of the TA repeats and the variation in the TA repeat length in nine serotype 15B and eight serotype 15C clinical isolates (positions 398 to 463 in the gene). The strain numbers are indicated in parentheses. Strain 981519 was identified by us as a serotype 15B strain. The premature stops in the cps15cM ORFs of serotype 15C strains are underlined.